Source Depth Estimation Based on Random Forest Approach Using Ocean Waveguide Data
DOI:
https://doi.org/10.53469/jrse.2025.07(01).15Keywords:
Passive localization, Normal mode, Random forest, Match field processingAbstract
In practice, the estimation of source localization based on matched field processing is significantly affected by environmental parameters, leading to the so-called mismatch problem. This paper models the sound source depth estimation problem as a classification issue in machine learning and discusses how the random forest method can be used to solve the depth estimation problem of sound sources. The paper uses the SWELLEX-96 sea trial environmental parameters and the Kraken model to generate ocean waveguide data received by a vertical line array at different depths of the sound source. After normalizing and extracting features from the generated ocean waveguide data, the random forest (RF) method is applied to estimate the depth of the sound source. The results indicate that the RF method is feasible for estimating the depth of sound sources.
References
Bucker, & Homer, P. (1976) Use of calculated sound fields and matched-field detection to locate sound sources in shallow water. Journal of the Acoustical Society of America, 59(2): 368-373.
Bucker, H. P., & Schey, P. W. (1985) Experimental test of acoustic localization in shallow water. Journal of the Acoustical Society of America, 77(S1): 253-262.
Frichter, G. M., Byrne, C. L., & Feuillade, C. (1990) Sector-focused stability methods for robust source localization in matched-field processing. The Journal of the Acoustical Society of America, 88(6):2843-2851.
Lei, Z., Yang, K., & Ma, Y. (2016) Passive localization in the deep ocean based on cross-correlation function matching. Journal of the Acoustical Society of America, 139(6): EL196- EL201.
Song, H. C., & Cho, C. (2017) Array invariant-based source localization in shallow water using a sparse vertical array. Journal of the Acoustical Society of America, 141(1): 183-188.
Michalopoulou, Z. H., Gerstoft, P., & Caviedes-Nozal, D. (2021) Matched field source localization with gaussian processes. JASA express letters, 1(6), 064801.
Chi, J., Li, X., Wang, H., Gao, D., & Gerstoft, P. (2019) Sound source ranging using a feed-forward neural network trained with fitting-based early stopping. Journal of the Acoustical Society of America, 146(3): EL258-EL264.
Liu, Y., Niu, H., & Li, Z. (2020) A multi-task learning convolutional neural network for source localization in deep ocean. The Journal of the Acoustical Society of America, 148(2):873-883.
Vera-Diaz, J. M., Pizarro, D., & Macias-Guarasa, J. (2021) Acoustic source localization with deep generalized cross correlations. Signal Processing, 187(2): 108169.
Wang, W., Wang, Z., Su, L., Hu, T., & Ma, L. (2020) Source depth estimation using spectral transformations and convolutional neural network in a deep-sea environment. The Journal of the Acoustical Society of America, 148(6): 3633-3644.
Liu, M., Niu, H., Li, Z., Guo,Y. (2024) Source depth estimation with feature matching using convolutional neural networks in shallow water. The Journal of the Acoustical Society of America, 155 (2): 1119–1134.
Murray J. (1996) The SwellEx-96 Experiment.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Linglin Shen, Xiangbo Sun
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
